The fundamental mechanism responsible for the transition of undifferentiated somatic cells (SCs) into gran- ulosa cells (GCs) during primordial follicle (PF) formation is poorly understood. Women with premature ovarian failure (POF), Turner Syndrome or inactivating mutation of FSH receptor are infertile and have defective and non-growing PFs. Therefore, the basic research on PF formation is necessary to understand the molecular basis of normal and defective folliculogenesis. We present compelling evidence that (1) bone morphogenic protein 2 (BMP2, a SC-derived ligand), GDF9 (an oocyte-derived ligand) and estradiol-17 (E2, an ovarian hormone) inter- act to promote PF formation in fetal ovaries, (2) the paracrine actions of GDF9 require BMP2 priming of SCs, and (3) inactivation of cadherin 2 (CDH2) interferes with PF formation.19 These important findings lead us to hypothesize that BMP2 and GDF9 spatio-temporally promote the transition of SCs into GCs and their assembly with the oocytes. E2 modulates the actions of BMP2 by upregulating the synthesis of BMP2 ligand and receptors. The specific aims (Fig. 1) are (1) Determine the mechanism of BMP2-mediated pri- mordial follicle (PF) formation. The objectives are, (1) To determine if BMP2 regulates the transition of SCs into GCs, (2) To determine if GDF9 action is required for or facilitates the transition of SCs to GCs, and (3) To determine if GDF9 facilitates oocyte-granulosa cell adhesion and assembly. We will knockdown BMP2 or GDF9 expression or block their actions or alter ALK2/3 activities in E15 hamster ovaries in vitro by small molecule inhibitors or shRNA. (2) Determine if estradiol-17 (E2) regulates BMP2-ALK3/ ALK2 system. The objective is to determine if E2 affects PF formation by regulating the expression and action of BMP2. We will knockdown or block BMP2 action or alter ALK2/3 activities by small molecule inhibitors or shRNA in hamster fetal ovaries, and in ovary recombinants to determine the mechanism of E2-induced SC to GC transition. (3) Analysis of gene networks affected by BMP2 or GDF9 during PF formation. The objective is to identify and analyze gene networks and corresponding pathways in SCs that are targeted by BMP2 or GDF9 during PF formation. We will use RNAseq analysis to determine differentially expressed genes (DEGs) in SCs of ovaries exposed to BMP2 or GDF9. High stringency cut off along with bioinformatics analysis will be used to select candidate genes for biological validation and will be prioritized based on their role in mesenchymal-epithelial transition (MET), in follicular development and functions, and the probability of their defects causing ovarian dysfunction. Biological validation will be accomplished using approaches outlined in Aim 1. This study will provide novel information to understand the basic mechanism responsible for the transition of undifferentiated SCs into GCs during PF for- mation, and may shed light to better understand the molecular basis of defects in PF formation.